Highly Active Photocatalyst of Cu2O/TiO2 Octahedron for Hydrogen Generation.

Heterojunction catalysts are attracting attention in the field of photocatalytic hydrogen generation for their effective light utilization and charge separation personalities. In this work, we report a simple and low-cost two-step solvothermal method for synthesizing Cu2O/TiO2 heterojunction catalysts with an octahedral morphology and a mean particle size of about 30 nm. It is found that the introduction of Cu2O astonishingly enhances the photocatalytic performance of TiO2. Under the condition of methanol acting as a sacrificial agent, the heterojunction with 0.19% Cu species shows an optimal hydrogen generation rate of 24.83 mmol g-1 h-1, which is nearly 3 orders of magnitude higher than that of the pristine TiO2 catalyst.

Introduction

Energy and environment are important to the progress and future of humanity. The continuous increase of world population together with the constant expansion of modern industry speeds up the existing energy consumption. Also, traditional energy utilization has little consideration on environment conservation and energy sustainability, which bring about imperious energy crisis and serious environmental issues.[1−3] Development of renewable clean energy is put forward for solving these problems at this crucial moment. Hydrogen, as a renewable and clean energy with unique advantages of high energy density and outstanding combustion efficiency, has been an important renewable clean energy and even been considered as the most hopeful energy in the 21st century.[4−7] Up to date, H2 can be produced by steam reforming of natural gas, electrolysis of water, photoelectrochemical or photocatalytic water splitting, and biological/microbial approaches. Among these methods, photocatalytic water splitting into H2 and O2 using solar energy and semiconductor photocatalysts has been considered as the most promising strategy to address the environment conservation and energy sustainability simultaneously.[8−11]

The overall water-splitting reaction requires a photocatalyst owning a more minus conduction band than H+/H2 reduction potential and a more positive valence band than O2/H2O oxidation potential. Representative photocatalysts are TiO2, ZrO2, KTaO3, SrTiO3, WO3, ZnS, CdS, MoS2, ZnO, SnO2, In2O3, C3N4, SrTiO3, BiVO4, Bi2MoO6, K4Nb6O17, and so on.[10,12−14] Unfortunately, most of these materials are not ideal candidates for water splitting because of their wide band gap, rapid recombination of photogenerated electron/hole pairs, or inferior photocorrosion resistance. For example, TiO2, as one of the most prominent photocatalysts for its high chemical stability and high photocatalytic activity under UV light, is also subject to its poor absorption of visible light and noticeable recombination of photogenerated carriers.[3,15−17] Various strategies have been proposed to solve these problems, for instance, by doping, metal loading, and/or constructing heterojunction.[18−21] Constructing heterojunctions by introduction of co-catalysts has been proved to be one of the most promising ways to improve the solar-to-chemical energy conversion (and chemical stability) of TiO2 and many other wide band gap (or unstable) semiconductors.[22−27] A good co-catalyst for constructing heterojunction should meet the following basic conditions: (i) small band gap (for the harvest of visible light); (ii) electron injection should be fast and efficient; and (iii) nontoxic, lowcost, and chemically stable. Besides, the energy position of the conduction band of TiO2 is very close to the H+/H2 reduction potential.[28−31] Therefore, the cocatalyst is generally required to possess a more negative conduction band than that of TiO2 for constructing a TiO2-based heterojunction. Cu2O is a semiconductor that meets all these requirements, and therefore the Cu2O/TiO2 heterojunction has attracted much attention.[32−37] The Cu species usually appeared in three forms: metal Cu, oxidation state CuI and CuII, which could exhibit concurrently in the composite and difficult to distinguish. Especially, the quantities were small in the composite. So far, little routes are available for fabricating such Cu2O/TiO2 catalysts that can both load with rare content of Cu and get pure Cu species of Cu2O.[38−40]

Herein, we successfully prepared the Cu2O/TiO2 catalyst with an octahedral structure and a nanoscaled size. The Cu species existing in the catalyst turned out to be Cu2O, and its content is no more than 1%. Furthermore, the composite catalyst exhibits superior photocatalytic performance when applied to generate hydrogen from a methanol solution. Along with the bonus of a high surface to volume, the Cu2O/TiO2 catalyst provides additional benefits of reuse to reduce the use cost sharply in the industrial-scale production.

Results and Discussion

Surface Morphology

Appropriate solvothermal reaction can convert protonated titanate and cupric acetate into titanium oxide and cuprous oxide nanoparticles, respectively. It is found that the morphology of the products was independent on the Cu/Ti ratio in this study. Figure shows the scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images of the representative sample T-3. The product consisted mainly of octahedron nanoparticles with a mean size of about 30 nm. On the basis of the knowledge of crystallography, these octahedron nanoparticles can be referred to anatase TiO2, and the eight exposed facets can be determined to be {101}. According to the Wulff construction, the octahedron structure enclosed by {101} facets is the thermodynamic equilibrium shape of anatase crystals.[41] The high-resolution TEM (HRTEM) image of the octahedron nanoparticles is displayed in Figure c and the corresponding fast Fourier transform (FFT) diffraction pattern is presented in Figure d. At least three sets of lattice fringes with spacings of 4.70, 3.51, 3.48, and 2.40 Å were observed, which correspond to (004), (101), (−101), and (103) planes of the anatase phase, respectively. Besides, as shown in Figure a,b, the nanoparticles were highly dispersed, which promised a high specific surface area. According to the nitrogen adsorption–desorption isotherm (Figure S1), the Brunauer–Emmett–Teller (BET) surface area of sample T-3 was as high as 101 m2/g.

Figure 1

(a) SEM image and (b) TEM image, (c) HRTEM image, and (d) corresponding FFT diffraction pattern of the selected area in (c) of catalyst T-3.

Crystalline Structure

Figure shows the X-ray diffraction (XRD) patterns of the Cu2O/TiO2 photocatalysts. For all the samples, only anatase TiO2 (JCPDS card no. 21-1272) was observed. It is found that the peak position of the anatase phase remained unchanged with the increasing copper content, which excluded the possibility of copper doping in the TiO2 lattice (or the doping concentration was very low). The difficult substitution of Ti by Cu can be attributed to their large mismatching of the valence state and ionic radius (0.605 Å for Ti4+, 0.73 Å for Cu2+, and 0.77 Å for Cu+). Furthermore, diffraction peaks of Cu2O and any other copper species were not observed in all the Cu2O/TiO2 samples, which may be due to the low content and the poor crystallinity of copper species because the inductively coupled plasma optical emission spectrometry (ICP–OES) (Table ) and scanning TEM–energy dispersed X-ray spectroscopy (STEM–EDS) (Figures S2 and S3) characterization confirm the presence of Cu. After calcining in argon atmosphere at 500 °C for 2 h, characteristic diffraction peaks of Cu2O were detected in T-6 with a copper content of about 4.4%, as shown in Figure S4. The emergence of Cu2O diffraction peaks after calcination in an inert gas demonstrated the existence of copper species, and copper existed in the form of Cu2O.

Figure 2

XRD patterns of TiO2 catalysts loaded with different contents of Cu2O. The standard diffraction pattern of anatase TiO2 (JCPDS card no. 21-1272) is provided at the bottom.

Table 1

Cu2O/TiO2 Catalysts Loaded with Different Contents of Cu

	Cu/Ti (at. %)
sample	designed	detected (ICP–OES)	H2 generation rate (mmol g–1 h–1)
T-0	0	0	0.03
T-1	0.02	0.0396	5.09
T-2	0.05	0.0722	9.93
T-3	0.2	0.1928	24.83
T-4	0.5	0.3752	20.89
T-5	1	0.9763	10.09

The structural property of the samples was further examined by Raman spectra. Anatase TiO2 is tetragonal and belongs to the space group (D419 = I41/amd), according to the factor group analysis, the fifteen optical modes have the irreducible representation Γopt = A1g (R) + 2B1g (R) + 3Eg (R) + B2u (ia) + A2u (IR) + 2Eu (IR).[42] As shown in Figure , the Raman peaks appeared at 144, 196, 396, 517, and 638 cm–1 can be definitely assigned to Eg(1), Eg(2), B1g(1), A1g + B1g(2), and Eg(3) modes of the anatase phase, respectively. Shift of the Raman peak was not detected, which further demonstrates that there was no doped copper in the TiO2 lattice.

Figure 3

Raman spectra for the Cu2O/TiO2 catalysts.

For further understanding the chemical environment of Ti and Cu, X-ray photoelectron spectrum (XPS) analysis was carried out. Figure shows the Ti 2p and Cu 2p spectra collected for the Cu2O/TiO2 photocatalysts. The corresponding peaks of the Ti 2p spin orbital appeared at 458.6 and 464.3 eV for Ti 2p3/2 and Ti 2p1/2, respectively, suggesting the presence of Ti4+ in all Cu2O/TiO2 photocatalysts. The highly symmetrical Ti 2p3/2 peak implied the absence of Ti3+, while its constant binding energy indicated the unchanged chemical environment. Figure b shows the Cu 2p spectra. Binding energies of Cu 2p3/2 and Cu 2p1/2 were found to be located at 932.7 and 952.5 eV, respectively, which can be assigned to Cu+. For Cu species, the shake-up satellite peak located at about 940–945 eV is characteristic for Cu2+ because of its d9 configuration in the ground state, while the d shell of Cu+ is completely filled (d10), and thus there was no satellite peak in the Cu 2p spectra for Cu2O.[32,43] Therefore, the absence of the shake-up satellite peak excluded the existence of Cu2+. On the basis of the analysis of the XPS and XRD results, we can conclude that Cu species exist in the form of Cu2O and there was no doped Cu atom in the TiO2 crystal lattice.

Figure 4

XPS spectra for the Cu2O/TiO2 catalysts. (a) Ti 2p and (b) Cu 2p.

UV–vis reflectance spectra were measured and converted from reflectance to absorption by the Kubelka–Munk method.[44,45] As shown in Figure a, additional visible-light absorption was induced and enhanced with increasing Cu2O content, which can be attributed to the absorption of Cu2O whose absorption band edge is 450–515 nm.[46,47] The enhanced absorption of visible light may be beneficial to improve the utilization efficiency of sunlight during the photocatalytic water-splitting process and thereby improve the photocatalytic performance. According to the band structure of the Cu2O/TiO2 heterojunction (Figure b), under the irradiation of visible light, photogenerated electrons transfer from the conduction band of Cu2O to that of TiO2 and then further transfer to the surface-active sites for H+/H2 reduction reaction. Obviously, this heterojunction structure can boost visible-light activity and reduce the recombination rate of photogenerated carriers.

Figure 5

UV–vis absorption spectra (a) and the energy band diagram (b) for the Cu2O/TiO2 catalysts.

The photocatalytic activity of the samples was investigated in a methanol solution under Xe lamp irradiation (Figure ) and visible light irradiation (Figure S5). Figure a shows the time-dependent hydrogen generation. For all the samples, the amount of produced H2 increased linearly with irradiation time. Figures b and S5b show the dependence of photocatalytic activity on the Cu2O content. It is found that the increase in copper loading (Cu/Ti ratio) results in increasing the hydrogen production up to 0.19% (i.e., sample T-3), beyond which a negative effect was dominated. The optimal photocatalytic hydrogen generation rate under the Xe lamp irradiation over sample T-3 was 24.83 mmol g–1 h–1, which was about 830 times higher than that of pristine TiO2 (0.03 mmol g–1 h–1). Similarly, the photocatalytic hydrogen generation activity under visible light (λ > 420 nm) irradiation over sample T-3 was also optimized by enhancing the hydrogen generation rate 17 times compared to pristine TiO2. A contrast was made to reveal the excellent performance for the Cu2O/TiO2 octahedron compared with other works as shown in Table S1. The improved reactivity for the Cu2O-loaded TiO2 photocatalysts can be attributed to both the enhancement of visible light absorption and the formation of the Cu2O/TiO2 heterostructure. As described above, the introduction of Cu2O with a narrow band gap enhances the light absorption of TiO2 (Figure a), which may induce visible-light photocatalytic activity. Furthermore, TiO2 is an n-type semiconductor, while Cu2O is a p-type semiconductor whose conduction band is more negative than that of TiO2. The contact of TiO2 and Cu2O nanoparticles constructs a p–n junction. Driven by the built-in electric field, the photogenerated electrons migrate from Cu2O to TiO2, while holes migrate in the opposite direction, decreasing the bulk recombination probability of photogenerated carriers. The electrons further migrate to the TiO2 crystal surfaces and react with water to generate H2, while the holes are consumed by the sacrificial agent. Obviously, the formation of heterojunction can greatly facilitate the separation of photogenerated carriers and improve the photocatalytic performance. Besides, the catalyst T-3 was evaluated ten cycles of reuse in methanol solution for H2 production under Xe lamp irradiation (Figure S6) and the photocatalytic activity did not decrease obviously. The used catalysts were annealed and characterized by XRD analysis as shown in Figure S7, and no observable change in the structure was detected. These observations indicated the high chemical stability of the photocatalysts.

Figure 6

Hydrogen yielded from a methanol solution over pristine TiO2 and Cu2O/TiO2 catalysts. Dependence of the amount of evolved hydrogen on the irradiation time (a) and the corresponding hydrogen generation rate (b).

Conclusions

In summary, we have successfully synthesized nano-sized Cu2O/TiO2 heterojunction catalysts with outstanding photocatalytic hydrogen generation performance from water splitting in the presence of the methanol sacrificial agent. The nano-sized Cu2O/TiO2 heterojunction catalysts have an octahedral morphology. It is found that the presence of Cu2O can enhance the photocatalytic performance of TiO2 by broadening the light absorption from the ultraviolet region to about 515 nm. The catalysts with 1.9% optimum copper content displays the highest hydrogen yield activity. The corresponding hydrogen generation rate from methanol solution is 24.83 mmol g–1 h–1, which is nearly 3 orders of magnitude better than that of the pristine TiO2. This work has demonstrated a simple and facile method to synthesize environment-friendly Cu2O/TiO2 heterojunction catalysts, which offers useful guidance for developing other transition-metal oxide composited with TiO2 photocatalysts using in energy conversion fields.

Experimental Section

Catalysts Preparation

A protonated titanate nanotube was used as the precursor for the synthesis of Cu2O/TiO2 heterogeneous photocatalysts. The H2Ti3O7 nanotube was prepared through a hydrothermal reaction procedure. In a typical process, 5 g of Degussa P25 was dispersed into 80 mL of 8 mol/L NaOH solution and magnetically stirred for 1 h. The mixture was transferred into a 100 mL autoclave and then heated at 180 °C for 10 h. After the hydrothermal reaction, the precipitate was collected and washed with diluted HNO3 solution (pH = 4) repeatedly until pH = 4, followed by washing with distilled water to neutral. The as-obtained protonated titanate product was further washed with absolute ethanol for several times and subsequently dried overnight at 60 °C in air.

A series of Cu2O/TiO2 catalysts were prepared by the alcohol-thermal method, cupric acetate monohydrate was dissolved into absolute ethanol to prepare 0.01 mol/L and 0.1 mol/L Cu(Ac)2 solution, and 5 g of H2Ti3O7 and a certain amount of 0.01 or 0.1 mol/L Cu(Ac)2 solution (0–65 mL) was added into ethanol and then magnetically stirred for 1 h. The mixture was transferred into a 100 mL autoclave for alcohol-thermal reaction under 150 °C for 20 h. The resulting precipitate was washed by water and alcohol repeatedly and dried overnight at 60 °C in air. The obtained Cu2O/TiO2 catalysts with different copper contents were named as T-0, T-1, T-2, T-3, T-4, and T-5, as shown in Table .

Catalyst Characterization

The crystal structure of the Cu2O/TiO2 photocatalyst was characterized by powder XRD on Rigaku MiniFlex 600 (40 kV, 30 mA) with a radiation source of Cu Kα (λ = 1.5418 Å). The data were collected in the range of 10°–80° at a scanning rate of 0.5°/min. The morphology of the catalyst was observed by field emission SEM using a Hitachi SU8010 with an applied voltage of 15 kV and TEM. The surface species analysis was characterized by XPS using a Thermo Fisher ESCALAB 250Xi electron spectrometer with a radiation source of monochromic Al Kα radiation and C 1s peak at 284.8 eV as internal standard. The Cu concentration was determined using ICP–OES (Ultima-2 ICP–OES analyzer, JY, France) and STEM–EDS. UV–vis absorption spectra were recorded on a Lambda 950 spectrophotometer (PerkinElmer, USA). Raman spectra were collected in the range of 100–900 cm–1 from a LabRAM HR instrument (Horiba Jobin-Yvon, France) using a 532 nm laser as the light source. The BET surface area of the catalyst was determined by N2 adsorption using an ASAP-2020 surface area analyzer.

Photocatalytic Activity Testing

In a typical procedure, 0.1 g of the Cu2O/TiO2 photocatalyst was dispersed in 100 mL 10 vol % aqueous methanol solution in a quarts reactor. The reactor was then equipped on an online photocatalytic activity evaluation system (CEL-SPH2N, CEAULight, China). With the existing of a low-temperature thermostat bath keeping the temperature of the reactor at 5 °C during the photocatalytic reaction, the aqueous methanol solution and catalyst would not be pumped as the airstream when the air in the system was removed by a pump and the yielded hydrogen was sampled during the entire photocatalytic process. The photocatalytic reaction began under the irradiation of a 300 W xenon lamp with or without a band pass filter (λ > 420 nm) accompanied with hydrogen yielded from the catalyst surface and detected by an appendant online chromatography.